In therapeutic areas with tough competition, it is challenging to establish a complete, confident understanding of a drug’s safety and efficacy profile. Any misunderstanding of a drug’s characteristics can derail the entire development program. Modeling and simulation, also known as model-based drug development, uses a number of techniques, such as population PK modeling, model-based meta-analysis, and PBPK modeling, to help gain a better understanding of a drug’s risk-benefit profile to increase the likelihood of regulatory success. In this blog post, I’ll discuss how I worked with my colleagues to help develop a novel combination therapy for obesity.
Demonstrating the safety of a treatment for obesity
A new combination product incorporating previously studied compounds presented high stakes when we first joined this project. Qsymia™ is a fixed-dose combination of topiramate XR and phentermine. This new candidate for the treatment of weight management faced added challenges in proving that it was safe, due to well-known issues encountered previously for a fenfluramine-phenteramine combination product (“fen-phen”) that was withdrawn from the market years earlier.
The therapeutic area—obesity—also faced regulatory skepticism as a condition meriting medical, rather than psychosocial, treatment, as well as additional recent withdrawals of approved drugs due to adverse events (ie, rimonabant). Additionally, the placebo effect for weight loss reached approximately three percent of body mass per year, high enough to rival many active drugs. Yet Qsymia showed great promise as an alternative to less-effective drugs, or costly and dangerous gastric surgery. It had blockbuster potential.
Providing a mechanistic understanding of drug effect on weight loss
We leveraged PK and PD data from Phase 2 and 3 studies and performed model based meta-analysis to tease out drug effect from placebo treatment, and ultimately confirmed the additive effects of topiramate and phentermine in obese patients. The low, mid, and full dose levels of Qsymia provided outstanding mean weight loss of 5.84%, 8.92% and 11.50%, respectively. Additive effects of Qsymia relative to monotherapy treatments phentermine and topiramate were modeled. Importantly, our model-based analysis of literature and trial data demonstrated that Qsymia did not display efficacy due to a drug-drug interaction between the two products. Overall, the combination of disease progression and exposure- response models provided a mechanistic understanding of the effect of Qsymia on weight loss.
Our auditors performed quality assurance on the population PK/PD report for the regulatory filing, and we were on hand to present findings and address questions at the FDA Advisory Board Meeting, as well as the regulatory hearing, when the sponsor submitted the NDA. FDA reviewers evaluated the dossier and ultimately approved Qsymia 20 to 2. The approval came neck-and-neck with that of its competitor, Belviq® (lorcaserin).
From developing a competitive strategy at Phase 1 through in-depth exposure-efficacy analyses, our model-based analysis offered compelling evidence to support the additive effect of the combination treatment on weight loss, and supported efficient creation of regulatory-compliant documents to support a safe and effective dosing scheme. We continue to apply the modeling framework to respond quickly to questions from regulatory authorities regarding dosing in other sub-populations, including patients with organ impairment, and to identify post-approval study requirements.
To learn more about this project, please check out this poster:
JF Marier, NH Gosselin, MS Mouksassi, N Kassir, S Yee, C Peterson, W Day. Population PK/PD Modeling of VI-0521, a Fixed-Dose Combination Product of Immediate-release Phentermine and Modified-release Topiramate for the Treatment of Obesity. PIII-78 S123-S124, 2012, Clinical Pharmacology & Therapeutics.
All information presented derive from public source materials.
My colleague, Colin Chang, recently gave a webinar where he demonstrated how we used the population PK modeling tool, Phoenix NLME, to optimize dosing of this drug in patients with hepatic impairment. I hope that you’ll view the on-demand recording and let me know what you think in the comments section!